INDUSTRY FIELD GUIDE
Sound recording studio take and track data for AI
A mix and an isolated microphone track can carry different editing histories.
NAICS 512240 · Sound Recording Studios
A mix and an isolated microphone track can carry different editing histories. The Library of Congress documents audio formats, and the Copyright Office distinguishes recordings from underlying works. A studio can show source lineage with an original scripted session and the right permissions. Library of Congress recommended formats · U.S. Copyright Office recording context
Record two short original takes with authorized participants, or use dummy filenames. Create a session list with take, track, sample rate and edit status. Give one export a mismatch and ask the reviewer which source it actually follows. Keep an unprocessed segment for evaluation. Document permission for any voice reuse explicitly, with separate treatment for uses outside the sample.
Do not use client sessions, performer likenesses, music or derived tracks unless clearly authorized. Owning the studio equipment does not grant client rights.
Describe an authorized recording sample
Original Fiund exercise
A mix filename cannot override the edit map's source take
Dummy session records describe two original scripted takes and an isolated microphone track. No audio, client session or performer identity is published.
| Record | Source | What it records |
|---|---|---|
| RS1 | Practice session list | Take A track Mic1 at 48000 samples/second; Take B track Mic1 at the same rate. Original A segment 0–2 seconds retained unchanged. |
| RS2 | Export map | MIX-X is cut from Take B/Mic1 at 1–3 seconds, with a documented trim. Filename says Take-A-final. |
| RS3 | Permission register | A and B scripted audio permitted for U; any separate voice-reuse purpose V requires a grant not supplied here. |
Question for the model
Which source does MIX-X actually follow, and can the filename or U permission establish an unprocessed Take A export or separate V use?
Show the reference answer and checks
RS2 explicitly maps MIX-X to Take B/Mic1 at 1–3 seconds. Its filename does not make it a Take A export, and the documented trim means it is not the retained unprocessed A segment in RS1. RS3 supports the specified U scope but leaves V ungranted. Keep take, track, sample rate, edit lineage and purpose-specific permission separate rather than relying on a final-file label.
- Selecting a source take from the export filename instead of its edit map.
- Treating permission for one scripted audio use as permission for a separate voice-reuse purpose.
Fields to preserve
- Session, take and track
- Identifies the original recording resource.
- Sample rate and source span
- Defines the timing basis for the export.
- Edit status and source map
- Distinguishes a trimmed derivative from an unprocessed segment.
- Use-specific voice permission
- Preserves the separate purpose boundary.
Test another case
Create an export combining A and B with two explicit source spans. Require piecewise lineage instead of a single take label for the whole mix.
Use dummy filenames or independently scripted recordings with explicit participant authority. Exclude client sessions, unrelated voices, music and protected derived tracks.
Discuss a contribution
Describe the sources you control, the task they could support and what has already been checked. Use the existing-data assessment for an archive or the collection brief for a new recording project. The contributor agreement defines Fiund’s licensing authority and compensation. A guide or fictional exercise does not establish an accepted collection or a sale.